#importing the desired modules

​

import pandas as pd

import numpy as py

import matplotlib.pyplot as plt

import plotly.express as px

​

cereal_df = pd.read_csv(r"C:\Solaris\MY COURSES\DATA ANALYTICS\DATA SETS\Cereal Data\cereal.csv")

​

pd.set_option('display.max_rows',None)   #Display all rows in the data frame

​

cereal_df.replace(-1,0,inplace = True)

​

​

​

​

cereal_df

#Statistical Summary of the  data

​

cereal_df.describe(include = 'all')

​

#Creating a graph depicting the various nutritional components of each cereal

​

nutrition_df = cereal_df[['name','protein','fat','sodium','fiber','carbo','sugars','potass','vitamins']]

nutrition_df.set_index('name',inplace = True)

​

​

fig1 = nutrition_df.plot(kind='barh',figsize = (20,500))

​

#Adding value labels in a chart

for i in range(0,8):

    plt.bar_label(fig1.containers[i],label_type = 'edge',padding = 15,fontsize = 20)

​

#increasing the size of the ticks

plt.xticks(fontsize = 20)

plt.yticks(fontsize = 20)

​

​

#adding labels

plt.xlabel('Quantity in grams',fontsize = 20)

plt.ylabel('Cereal',fontsize = 20)

​

#Increasing the fontsize of legend

plt.legend(fontsize = 20)

​

​

​

​

# Determing the foods with the highest calories.

​

calories_df = cereal_df[['name','calories']]

​

calories_df

​

fig_calories = px.scatter(calories_df, x= "calories",y = "name")

​

fig_calories.update_layout(

                           title = 'CEREALS AND THEIR CALORIE CONTENT',

                           font_family = "Georgia",

                           title_font_family = "Times New Roman",

                           title_font_color = "Blue",

                           title_x =0.5,

                           xaxis_title = "Calories",

                           yaxis_title = "Cereal Name")

​

fig_calories.update_xaxes(title_font_color = "Blue",title_font = {'size':15},title_standoff = 25)

fig_calories.update_yaxes(title_font_color = "Blue",title_font = {'size':15},title_standoff = 25)

​

fig_calories.show()

​

# Which manufacturer produces the most popular cereals?

#For this we will take the average of the ratings for each manufacturer and create a pie chart

​

​

popular_df = cereal_df[['mfr','rating']]

​

group = round(popular_df.groupby('mfr').mean(),2)

​

group.reset_index(inplace = True)

figure = px.pie(group, values = 'rating',

                names = 'mfr',

                title = "Manufacturers Vs Ratings",

                hole = 0.3,

                color_discrete_sequence = px.colors.sequential.Plasma_r)

​

figure.update_layout(title_x = 0.5)

figure.update_traces(textinfo = 'label + value')

figure.show()

​

#From the pie chart it can be deducted that N manufacturer is the king of the market.

# N has the highest rating(average of rating of it's products) than the other manufacturers.

​

​

​

# Creating a sunburst chart to easily identify the top-rated and the bottom-rated cereals

# The chart also helps in easy visual identification of each manufacturer's popular and least favourite cereals

​

sun_fig = px.sunburst(cereal_df, path = ['mfr','type','name'],color = 'rating',color_continuous_scale = 'RdBu')

sun_fig.show()

​

​

# All of N manufacturer's cereals are rated highly.

# All of G manufacturer's cereals belong to the lower end of the rating scale.

# While all manufacturers produce cereal of type C,

# Manufacturer's N and Q are the only ones producing additional cereals of type H.

​

#Using a trendline to affirm that rating and calories have an inverse relationship

#Fewer the calories, higher the rating of the cereal

​

import seaborn as sns

sns.regplot(x = 'calories',

            y='rating',

            data = cereal_df,

            scatter_kws = {'color':'orange'})

​

sns.set(rc = {'figure.figsize':(10,5)})

plt.ylim(0,)

​